Overview

We follow a single research methodology across all industries. Our scope is locked first, and data is collected next. Validation is enforced. Models are reconciled. KPIs are computed. Lineage is published. The outcome is fully auditable and consistently repeatable.

What is fixed upfront

  • Scope, taxonomy, units (value/volume), seven regions and thirty countries
  • Base year and time granularity

How content is organized

  • B2B and B2C collection & validation
  • Mathematical models with equations and checks
  • Sources, cadence, and FAQs

Country → Region → World (three-fold roll-up)

We start with vendor-tracked country numbers and size the remaining tail. Supply and demand are analyzed at the country level. Regions are then reconciled against disclosed totals. The final roll up yields a global view where capacity and consumption are brought into balance.

Country

  • Shipments counted once at final brand; subsidiary inventory excluded until channel entry
  • Tail completed with tier model (mean revenue × vendor counts)
  • Supply–use identity checked against production, trade, stocks

Region

  • Sum of countries; constrained least squares nudges to match filings bands
  • Countries kept inside price and capacity corridors

World

  • Regions sum; capacity & technology arcs tested
  • Vintage stability tracked (sMAPE/MASE; Diebold–Mariano)

Measurement

Definitions are precise to shorten audits and prevent ambiguity.

Shipments & ASP

  • Shipments counted once at final brand; re-exports reassigned to final use when determinable
  • ASP at distributor level; includes freight/insurance/import duties in vendor/channel pricing; excludes POS taxes; outliers normalized
  • Value = Units × ASP; Installed base = cumulative shipments − retirements

Capacity & FX

  • Capacity ≈ Production / φ (φ ≈ 0.80–0.90)
  • History in current USD; forecasts at constant FX unless scenario work requires otherwise

Factor-100

100 drivers in 10 families, with weights published on lineage cards. Caps/floors protect against overfit and implausible jumps.

  1. Real GDP growth
  2. GDP per capita
  3. CPI (headline)
  4. PPI / WPI
  5. Unemployment rate
  6. Consumer confidence index
  7. Business confidence / PMI composite
  8. Gross fixed capital formation
  9. Exchange rate (USD pair)
  10. Policy rate / yield (10Y)
  1. HS‑line import volume
  2. HS‑line import value
  3. HS‑line export volume
  4. HS‑line export value
  5. Net trade balance (qty)
  6. Net trade balance (value)
  7. Transit time / port dwell
  8. Freight index (e.g., SCFI)
  9. Tariff / NTB dummy
  10. Customs clearance time
  1. Feedstock price benchmark
  2. Spot commodity (domain‑specific)
  3. Energy price (electricity)
  4. Natural gas price
  5. Crude / refined price
  6. Ocean freight rate
  7. Truckload/LTL index
  8. Wage index (manufacturing)
  9. Distributor margin index
  10. Retail price index (category)
  1. Industrial production (2‑digit)
  2. Capacity (nameplate)
  3. Capacity utilization
  4. Plant uptime / outages
  5. Capex announcements
  6. New line installs
  7. OEM order backlog
  8. Maintenance shutdowns
  9. Yield / scrap rate
  10. OEE proxy
  1. New standard enactment
  2. Compliance deadline dummy
  3. Subsidy/incentive on‑off
  4. Import/export restriction
  5. Safety recall count
  6. Environmental limit change
  7. Labeling/UDI mandate
  8. Tax/VAT change
  9. Trade remedy (AD/CVD)
  10. Government procurement rule
  1. Patent filings (class)
  2. Standards ratification
  3. Installed base of enabling tech
  4. Software release cadence
  5. Cloud/edge adoption
  6. Sensor/telemetry penetration
  7. Yield/efficiency improvement
  8. Learning curve factor
  9. Open‑source project velocity
  10. Cyber incident index
  1. Approvals (FDA/EMA/PMDA)
  2. UDI/device registry count
  3. Clinical trials active
  4. Procedure volumes
  5. Bed occupancy
  6. Payer schedule update
  7. Reimbursement coverage
  8. HTA outcomes
  9. Adverse event rate
  10. Hospital capex index
  1. Search interest index
  2. App installs / MAU
  3. SRAP (engagement velocity)
  4. Promo Intensity Index
  5. Retail Heat Index
  6. Store network delta
  7. BOPIS/delivery windows
  8. Returns ratio
  9. Average review rating
  10. Creator‑tier mix
  1. Carbon intensity
  2. Renewables share
  3. Water withdrawal
  4. Waste/recycling rate
  5. Scope 3 disclosure rate
  6. ESG policy score
  7. Ecolabel penetration
  8. Material restriction flags
  9. Compliance audit pass rate
  10. Supplier code adoption
  1. Geopolitical risk index
  2. Disaster incidence
  3. Strike/lockout days
  4. Cyberattack severity
  5. Currency crisis dummy
  6. Sanctions list changes
  7. Pandemic/public health alerts
  8. Recall event score
  9. Credit stress spread
  10. Supply outage severity

Worked examples

Two simple calculators are shared for transparency—CAGR gap-fill and BED driver blend.

CAGR gap-fill

Use the growth between 2017 and 2019 to fill 2018 (log-linear assumption, documented on lineage).

$\\text{CAGR}=\\left(\\frac{V_{2019}}{V_{2017}}\\right)^{1/2}-1 \\,;\\; \\hat V_{2018}=V_{2017}(1+\\text{CAGR})$

Estimated 2018:

BED driver blend

Blend utilization, orders, and energy-cost drivers with weights that sum to 1. See the next-year value in one click.

$\\text{CAGR}=\\left(\\frac{V_{2019}}{V_{2017}}\\right)^{1/2}-1 \\,;\\; \\hat V_{2018}=V_{2017}(1+\\text{CAGR})$
Vt
g1%
w1
g2%
w2
g3%
w3

Vt+1:

Sources, lineage, and QC

Every chart ships with a lineage chip, confidence tier, and caveat. Short text near numbers; long annex on request.

Field What is shown
Sources UN Comtrade, ITC, Eurostat/PRODCOM, US Census/BEA/BLS, OECD, IEA/EIA, USGS, FAOSTAT/USDA, national stats, tender boards, FDA/EMA/PMDA/CDSCO, ClinicalTrials.gov/WHO ICTRP, OICA/ACEA/SIAM
Access Public / Licensed / Consent
Transforms Unit conversions, currency basis, seasonality method
Confidence A / B / C tier with reason
Caveats Short and specific

Measurement Rules (formal)

$V=U\\times ASP \\quad ; \\quad IB_t = IB_{t-1} + Ship_t - Retire_t$
$Cap \\approx Q/\\phi,\\ \\phi\\in[0.80,0.90] \\quad ; \\quad C = Q + M - X \\pm \\Delta S$
  • Shipments are attributed to the brand on the final product; OEM hand-offs are not double-counted.
  • ASP is distributor-level; excludes POS taxes; outliers normalized; regional weights reflect vendor mix.
  • History in current USD; forecasts at constant FX unless scenario requires an alternative path.

Country Formulas & Gap-fills

$V_{v,c,t}=U_{v,c,t}\\cdot P_{v,c,t}\\ ;\\ V_{c,t}^{tracked}=\\sum_v U_{v,c,t}P_{v,c,t}\\ ;\\ V_{c,t}^{tail}=\\sum_{k\\in T}\\bar R_{k,c,t}N_{k,c,t}$
$Q_{c,t}+M_{c,t}-X_{c,t}\\pm\\Delta S_{c,t} \\approx C_{c,t} \\approx (V_{c,t}^{tracked}+V_{c,t}^{tail})/\\bar P_{c,t}$

Gap-fills

CAGR: $\\hat V_{t1+1} = V_{t1}\\left(\\frac{V_{t2}}{V_{t1}}\\right)^{1/(t2-t1)}$
Driver blend: $\\hat V_t = V_{t-1}\\prod_j (1+w_j g_{j,t})$, $\\sum w_j=1$

Regional Alignment (soft constraints)

$V_{r,t}=\\sum_{c\\in r}(V_{c,t}^{tracked}+V_{c,t}^{tail})$
$\\min \\sum_r\\left[\\dfrac{V_{r,t}-\\hat V_{r,t}^{filings}}{\\sigma_r}\\right]^2 \\ \\text{s.t.}\\ \\sum_{c\\in r}V_{c,t}=V_{r,t},\\ V_{c,t}\\ge0$

Countries are nudged within feasible price/capacity corridors; filings define soft bands, not hard overwrites.

Forecast Engines

Method 1 — Regression

$\\ln V_{c,t}=\\alpha+\\sum_j \\beta_j \\ln D_{j,c,t-\\ell_j}+\\gamma Z_{c,t}+\\mu_c+\\tau_t+\\varepsilon_{c,t}$
  • Diagnostics: stationarity (ADF/PP), multicollinearity (VIF), heteroskedasticity-robust errors (HAC/cluster), structural breaks (Chow/Bai–Perron)
  • Out-of-sample: sMAPE, MASE; Diebold–Mariano tests vs naïve
Forecast Engines

Method 2 — Driver growth-rate blend

$\\widehat V_{t+1}=V_t\\cdot \\prod_j(1+w_j g_{j,t+1}),\\ \\sum w_j=1$
Vt
g1%
w1
g2%
w2
g3%
w3

Vt+1:

Accuracy & QC

  • MAPE, sMAPE, MASE; correlation for blends; DM tests for forecast comparisons
  • Residual checks: ACF/PACF, Ljung–Box, ARCH; parameter drift (CUSUM)
  • Reconciliation: supply–use closure; price and capacity corridors; FX consistency
  • Governance: dual-analyst sign-off; assumptions defended; change log recorded
Chat With MaRIA

Chat With

MaRIA

Hi There, I am MaRIA, Your Market Research Intelligence Assistant. How may I help you?